The emergence of the New Testament as a uniquely authoritative body of texts has always constituted an important strand in the early history of the Christian Church. David Dungan contends that there was no canon of sacred Scripture before the fourth century because ‘so far no official or ecclesiastical body had met to determine which writings to accept and which to reject in Holy Scripture’ (p. 30). He holds that such a canon (by which he means a closed and defined list) was established as a result of the ‘strategy in defence of the Catholic Scriptures’ mounted by Eusebius of Caesarea and that ‘Constantine reshaped Catholic Christianity and its Scriptures’. Such a thesis sounds radical and innovatory, and Dungan indeed presents it as such. The thesis is in fact neither original nor as radical as it seems in view of its extremely narrow definition of what constitutes a canon. Somewhat surprisingly, Dungan never discusses the fragmentary text published by L. A. Muratori in 1740, commonly known as ‘the Muratorian fragment’ and sometimes called ‘the Muratorian canon’, which is normally dated c. 200. Nor does he refer to G. M. Hahneman's The Muratorian fragment and the development of the canon (Oxford 1992) or to the long and thoughtful review of Hahneman's monograph by Everett Ferguson (JTS n.s. xliv [1993], 691–7), which makes the necessary distinction (that Dungan blurs) between a canon and a ‘closed canon to which nothing could be added’. On the substantive issue which is the focus of Dungan's book, David Trobisch, The first edition of the New Testament (Oxford 2000), which was originally published in German in 1996, has advanced a strong argument for believing that the manuscripts of the New Testament all derive ultimately from an edition of the whole Bible put together in the middle of the second century. Any scholarly treatment of Dungan's subject needs to begin by explaining why Trobisch is wrong – if indeed he is wrong.